keywords Anopheles gambiae, Culex quinquefasciatus, insecticide-treated nets, pyrethroids, resistance, Tanzania

Transcription

1 Tropical Medicine and International Health doi: /j x volume 12 no 9 pp september 2007 Efficacy of pyrethroid-treated nets against malaria vectors and nuisance-biting mosquitoes in Tanzania in areas with long-term insecticide-treated net use Manisha A. Kulkarni 1,2, Robert Malima 2,3, Frank W. Mosha 2, Shandala Msangi 4, Ezra Mrema 5, Bilali Kabula 6, Boniface Lawrence 4, Safari Kinung hi 6, John Swilla 6, William Kisinza 6, Manfred E. Rau 1, Jane E. Miller 7, Joanna Armstrong Schellenberg 5,8, Caroline Maxwell 2,3,8, Mark Rowland 8, Stephen Magesa 2,3 and Chris Drakeley 2,8 1 Department of Natural Resource Sciences, McGill University, Ste Anne de Bellevue, QC, Canada 2 Joint Malaria Programme, Moshi, Tanzania 3 National Institute for Medical Research, Amani Research Centre, Muheza, Tanzania 4 Tropical Pesticides Research Institute, Arusha, Tanzania 5 Ifakara Health Research & Development Centre, Ifakara, Tanzania 6 National Institute for Medical Research, Dar es Salaam, Tanzania 7 Population Services International, Dar es Salaam, Tanzania 8 London School of Hygiene & Tropical Medicine, London, UK Summary objective To measure pyrethroid susceptibility in populations of malaria vectors and nuisance-biting mosquitoes in Tanzania and to test the biological efficacy of current insecticide formulations used for net treatment. methods Anopheles gambiae Giles s.l., An. funestus Giles s.l. and Culex quinquefasciatus Say were collected during three national surveys and two insecticide-treated net (ITN) studies in Tanzania. Knockdown effect and mortality were measured in standard WHO susceptibility tests and ball-frame bio-efficacy tests. Test results from 1999 to 2004 were compared to determine trends in resistance development. results Anopheles gambiae s.l. and An. funestus s.l. were highly susceptible to permethrin (range %) and deltamethrin (consistently 100%) in WHO tests in 1999 and 2004, while Culex quinquefasciatus susceptibility to these pyrethroids was much lower (range 7 100% and 0 84% respectively). Efficacy of pyrethroid-treated nets was similarly high against An. gambiae s.l. and An. funestus s.l. (range %) while efficacy against Cx. quinquefasciatus was considerably lower (range 2 100%). There was no indication of development of resistance in populations of An. gambiae s.l. or An. funestus s.l. where ITNs have been extensively used; however, susceptibility of nuisance-biting Cx. quinquefasciatus mosquitoes declined in some areas between 1999 and conclusion The sustained pyrethroid susceptibility of malaria vectors in Tanzania is encouraging for successful malaria control with ITNs. Continued monitoring is essential to ensure early resistance detection, particularly in areas with heavy agricultural or public health use of insecticides where resistance is likely to develop. Widespread low susceptibility of nuisance-biting Culex mosquitoes to ITNs raises concern for user acceptance of nets. keywords Anopheles gambiae, Culex quinquefasciatus, insecticide-treated nets, pyrethroids, resistance, Tanzania Introduction A key component of the global strategy to reduce malaria is the promotion of insecticide-treated nets (ITNs). In several regions of Tanzania, social marketing has proven effective as a means of increasing ITN coverage (Premji et al. 1995; Schellenberg et al. 2001; Hanson & Worrall 2002), with resulting impacts on malaria morbidity and mortality similar to those observed in randomized trials in other areas of Africa (Lengeler 2004). However, the widespread development of resistance to pyrethroid insecticides in malaria vectors, now recorded from West Africa (Elissa et al. 1993; Chandre et al. 1999; Awolola et al. 2002; Etang et al. 2003), East Africa ª 2007 Blackwell Publishing Ltd 1061

2 (Vulule et al. 1994; Ranson et al. 2000; Stump et al. 2004) and South Africa (Hargreaves et al. 2000), raises concern over the sustainability of ITNs for malaria control. Both public health and agricultural use of pyrethroids may contribute to the development of resistance in mosquito populations. In Kenya, use of permethrin-treated nets was associated with reduced permethrin susceptibility in An. gambiae s.s. (Vulule et al. 1994), while in South Africa indoor spraying with deltamethrin resulted in pyrethroid resistance in An. funestus (Hargreaves et al. 2000). Agricultural use of pyrethroids, primarily in cottongrowing areas, has contributed to selection for resistance in An. gambiae s.s. in Cote d Ivoire, Benin and Burkina Faso (Akogbeto & Yakoubou 1999; Chandre et al. 1999; Diabate et al. 2002). It is unclear how resistance will affect the level of malaria control achieved by ITNs, as this may vary with the molecular mechanism(s) of resistance present in the vector population (Chandre et al. 2000; Corbel et al. 2004) and different results have been found in ITN trials in areas with resistance in West Africa (Henry et al. 2005; N Guessan et al. 2007); however, the potential threat posed by development of resistance in vector populations has long been recognized (Curtis et al. 1998). While randomized controlled trials across Africa demonstrated high ITN efficacy (Lengeler 2004), it was uncertain whether the high small-scale coverage achieved by free delivery of nets in a trial setting could be replicated on a larger scale and in a more easily sustainable way. The social marketing approach was developed as a step towards a feasible and sustainable large-scale delivery strategy. The success of this approach was then used to plan for nationalscale initiatives. Here we define the term social marketing to mean using approaches from commercial marketing for a social, rather than financial, gain (Andreasen 1995), recognizing the differences in social marketing strategies between different organisations and over time. An early ITN effectiveness study in Tanzania, the Bagamoyo Bednet Project (BBNP), was conducted in 13 villages in the Bagamoyo district from 1991 to 1994 (Makemba et al. 1995; Premji et al. 1995). While this study used a relatively simple community-based distribution system for nets (Makemba et al. 1995), the larger Kilombero and Ulanga Treated Net Project (KINET), , developed a network of wholesale and retail sales agents to supply nets to all 109 villages in the Kilombero and Ulanga districts of Tanzania (Schellenberg et al. 1999). Social marketing was applied on an even larger scale in the Social Marketing of Insecticide-Treated Nets (SMITN) project, which was initiated in 1998 in Dodoma, Morogoro, Mtwara and Dar es Salaam regions, and expanded in 2000 to cover all regions of Tanzania (Hanson & Worrall 2002), becoming SMARNET from 2002 onwards. Like KINET, the SMITN project used a combination of public and commercial sector distribution systems. Other ITN projects that have been operating in the Tanga region of Tanzania since 1987 have used a system of free net delivery and regular re-treatment. A timeline of ITN projects in Tanzania is presented in Figure 1. The number of ITNs sold in Tanzania through the SMITN and SMARTNET projects gives an indication of the evolution of the standing crop of ITNs in the country. Assuming an average 5-year lifespan of a net and annual net re-treatment, Figure 2 shows that the standing crop of ITNs in Tanzania (estimated from cumulative net sales) has increased steadily from 1 million in 2002 to more than 9 million in Given the potential selection pressure for insecticide resistance from exposure of mosquitoes to ITNs or agricultural insecticides, this study aimed (1) to monitor the insecticide susceptibility status of local mosquito populations in areas of Tanzania where ITN projects have operated or where there is a relatively high use of pyrethroids for agricultural purposes and (2) to test the efficacy of the pyrethroid insecticides currently marketed for net re-treatment. The major vectors of malaria, Anopheles gambiae s.l. and An. funestus s.l., and the widespread nuisance-biting mosquito and vector of Bancroftian filariasis, Culex quinquefasciatus, were included in the study. Entomological data were compiled from three national surveys to assess the level and distribution of mosquito pyrethroid susceptibility in Tanzania. Two surveys were commissioned by Population Services International (PSI), Tanzania, in 1999 and 2004, which provided data on mosquito insecticide susceptibility and net bio-efficacy from areas of Tanzania with increasing ITN coverage. In a National Malaria Control Programme (NMCP)-commissioned survey provided additional insecticide susceptibility data from eight regions of Tanzania. To further assess the impact of long-term ITN use, entomological data from two major ITN study areas, in the Kilombero valley of southern Tanzania and the Tanga region of northern Tanzania, were interpreted in the context of trends in ITN coverage and resistance development. Methods Figures 3 and 4 show the regions of Tanzania where mosquitoes were collected in For the PSI-commissioned survey four sites were visited between August and September 1999 in the four regions of Tanzania where the Social Marketing of Insecticide Treated Nets (SMITN) project introduced ITNs in Both urban and rural areas were included: Mvumi and Igunguli villages, 1062 ª 2007 Blackwell Publishing Ltd

3 Community-based ITN delivery: Bagamoyo Bednet Project (BBNP) Social marketing ITN delivery: Kilombero and Ulanga Net Project (KINET) Social Marketing of Insecticide-Treated Nets Project (SMITN) / SMARTNET 4 regions Nationwide Project-based (free) ITN delivery: Tanga region village-based ITN studies Figure 1 Timeline of insecticide treated net (ITN) distribution in Tanzania through community-based, social marketing and project-based (free) delivery. Solid arrows represent project implementation and duration; dotted arrows represent continued ITN distribution through established social marketing systems Figure 2 Sales of nets (bundled with insecticide treatment kits) in Tanzania since 1998 by Population Services International (PSI) in the SMITN and SMARTNET projects. Annual net sales are shown in solid columns; cumulative net sales (adjusting for average 5-year lifespan of a net) are shown in striped columns. The cumulative number of nets is an indication of the standing crop of insecticide-treated nets in Tanzania in a given year. Number of bundled nets Year 2006 Dodoma region; Changarawe village and Morogoro town, Morogoro region; Mtoni kwa Azizi Ali area, Dar es Salaam region; Mikindani village and Mtwara region. Njagi village in the Kilombero district of Tanzania was included in the KINET study area, where between 1997 and 1999 ITN coverage of infants increased from less than 10% to more than 50% through implementation of social marketing strategies (Schellenberg et al. 2001). The study area is described in detail by Schellenberg et al. (1999). ª 2007 Blackwell Publishing Ltd 1063

4 Figure 3 Map of Tanzania showing susceptibility of Anopheles spp. to permethrin (0.75%) and deltamethrin (0.05%) in WHO susceptibility tests, Percentage mortality is represented by the shaded portion of the pie chart. Figure 4 Map of Tanzania showing susceptibility of Culex quinquefasciatus to permethrin (0.75%) and deltamethrin (0.05%) in WHO susceptibility tests, Percentage mortality is represented by the shaded portion of the pie chart ª 2007 Blackwell Publishing Ltd

5 Mn gaza village is situated in the Muheza district of Tanzania. For the Tanga Study ITNs were first introduced in 1987 (Curtis et al. 1998), since which time regular insecticide re-treatment and replacement of torn nets has maintained 70 90% ITN coverage in this village (Maxwell et al. 2006) Study sites For the PSI-commissioned survey the four SMITN project study sites as well as two other sites where long-term ITN projects have operated (KINET, Ifakara and BBNP, Bagamoyo) were revisited between November 2004 and January Urban and rural areas were included in the survey: Mvumi village, Dodoma region; Mlali, Mindu and Lupiro villages, and Morogoro and Ifakara towns, Morogoro region; Keko and Gerezani areas, Dar es Salaam region; Mikindani village, Mtwara region; Mangesani and Kigongoni vitongoji (subvillages) of Bagamoyo, Pwani region. For the NMCP-commissioned survey, nine sites across Tanzania were included in a resistance monitoring survey commissioned by the National Malaria Control Programme in : Mvomero and Ifakara, Morogoro region; Kyela, Mbeya region; Magu, Mwanza region; Arusha and Babati, Arumeru region; Manolo village, Tabora region; Lower Moshi, Kilimanjaro region and Ilala, Dar es Salaam. Mosquito collection Early morning indoor-resting catches were carried out in all locations with the exception of Mn gaza village. Freshly blood-fed, female Anopheles and Culex mosquitoes were aspirated from their resting sites on the walls and other surfaces inside houses and transported to a field laboratory or other suitable test location for susceptibility tests and bio-efficacy tests. In Mn gaza village in 1999 larvae were sampled and reared to the adult stage for testing. Susceptibility tests Standardized test papers impregnated with diagnostic concentrations of insecticides were used for susceptibility tests in all surveys and studies. WHO-recommended diagnostic concentrations of pyrethroids were increased in 2000 following a multi-centre trial (WHO 1998), therefore susceptibility tests carried out in 1999 or earlier used the former diagnostic concentrations of permethrin (0.25%) and deltamethrin (0.025%), while the increased concentrations of permethrin (0.75%) and deltamethrin (0.05%) were used in later surveys. DDT (4.0%) was tested in addition to pyrethroids in the NMCP-commissioned survey. Susceptibility of mosquitoes to diagnostic concentrations of permethrin, deltamethrin and DDT was assessed using standard WHO (1992) methods. Control and test replicates with batches of mosquitoes were carried out simultaneously. The number of mosquitoes knocked down was recorded at 5-min intervals during a 1-h exposure period. Mortality was scored after a 24-h holding period, during which time mosquitoes were provided with a 10% sugar solution. Resistance was defined according to WHO guidelines which state that % mortality indicates susceptibility, 80 97% indicates the possibility of resistance that needs to be confirmed, and <80% indicates resistance (WHO 1998). Bio-efficacy tests Bio-efficacy tests conducted during the KINET study and the 1999 and 2004 PSI commissioned surveys used standard polyester bednets (KINET study: Siamdutch Ltd., Bangkok, Thailand; PSI surveys: A to Z Textiles, Arusha, Tanzania) that were treated with pyrethroid formulations according to the instructions enclosed within the insecticide treatment kit (Ngao TM, PSI, Dar es Salaam, Tanzania). Deltamethrin (KO-tab TM, Bayer Environmental Science, Johannesburg, South Africa; 25 mg ai m 2 ) was tested in both 1999 and In addition, the newer alphacypermethrin formulation (Fendona TM BASF, Gravelines, France 6% SC; 40 mg ai m 2 ) was tested in Mosquitoes were exposed to a piece of ITN stretched over a wire ball-frame of approximately 15 cm diameter (WHO 1998). Ten mosquitoes were introduced into the test chamber and their exposure timed for 3 min; the number knocked-down was recorded at 3, 20, 30 and 60 min. Mosquitoes were provided with a 10% sugar solution and mortality was scored after 24 h. The ballframe was repositioned after each replicate to allow for variation in insecticide coverage; control replicates using an untreated net were run simultaneously. Statistical analysis Percent mortality was corrected by Abbott s formula when mortality in control replicates was >5% (Abbott 1925). Tests where control mortality exceeded 20% were excluded from analysis. Time taken for 50% knockdown of mosquitoes (KT 50 ) and 95% confidence intervals were determined by probit analysis using the computer program PoloPlus (Version 1.0, LeOra Software) (Finney 1971). Lethal dose ratios (LD 50 ) were calculated using estimates of the slopes and intercepts of two probit lines to determine ª 2007 Blackwell Publishing Ltd 1065

6 the relative susceptibility of mosquito populations to different insecticide treatments or in different years. Probit lines were tested for equality by chi-squared (v 2 ) tests of slopes and intercepts sampling sites is shown in Figures 3 and 4, indicating the level of pyrethroid susceptibility of Anopheles spp. and Cx. quinquefasciatus, respectively, in each area. Results Mosquito collection Seasonal fluctuations in mosquito densities made it difficult to consistently collect large numbers of mosquitoes from all sites. During the study period, heavy rains in the coastal areas of Bagamoyo and Mtwara and drought conditions in Dodoma hampered efforts to collect large numbers of Anopheles spp. Dar es Salaam was visited twice during the PSI-commissioned survey, but very few (<10) adult Anopheles were collected from households despite the visible presence of larvae. Efforts to rear fieldcollected larvae to the adult stage for testing in Dar es Salaam were not successful. Survivors and dead mosquitoes from susceptibility tests and bio-efficacy tests in the PSI-commissioned survey were stored desiccated in microcentrifuge tubes with silica gel and refrigerated when possible. Unfortunately, due to technical difficulties during fieldwork leading to poor preservation of specimens it was not possible to perform molecular analysis to identify An. gambiae s.l. sibling species or to test for knockdown resistance (kdr) alleles. Susceptibility tests Table 1 lists the results of susceptibility tests done in 1999 and 2004 from PSI- and NMCP-commissioned surveys. In the majority of sites An. gambiae s.l. was the only malaria vector species tested, while An. funestus s.l. was tested in Morogoro and Mtwara. Susceptibility >85% of An. gambiae s.l. and An. funestus s.l. to pyrethroids was observed in all areas. Complete susceptibility of these Anopheles species to deltamethrin was observed; however, slightly reduced levels of susceptibility to permethrin were noted in Dodoma (90%) and Morogoro (87%) in 1999 and in Ifakara (97%), Arumeru (91%) and Moshi (96%) in Lower susceptibility in 1999 may reflect the lower diagnostic concentration used at this time, since complete susceptibility was observed in tests using the increased dosage in Anopheles spp. susceptibility to DDT (4.0%) was generally high (98 100%), with slightly reduced mortality in Ifakara (93%) and Tabora (95%). Cx. quinquefasciatus susceptibility was generally much lower in all sites, ranging from 7% to 100% for permethrin (0.75%) and 49 84% for deltamethrin (0.05%). A map of Bio-efficacy tests Mortality after 3-min exposure to alphacypermethrin- and deltamethrin-treated nets in 1999 and 2004 is presented in Table 2. In general pyrethroid-treated nets were highly efficacious against An. gambiae s.l. and An. funestus s.l., although between-replicate variability in tests in Ifakara in 2004 decreased the calculated mean mortality. In 2004, no significant difference in efficacy was observed between the pyrethroids in Dodoma (P > 0.05) while in Mtwara deltamethrin produced more rapid knockdown and higher mortality than alphacypermethrin (P < 0.05). Susceptibility of Cx. quinquefasciatus was lower than that of Anopheles spp. in all sites in 2004; however, higher levels of mortality were achieved with deltamethrin in Bagamoyo and Mtwara compared to alphacypermethrin (P < 0.05). Changes over time Data from the 1999 and 2004 PSI-commissioned surveys on susceptibility of Cx. quinquefasciatus to deltamethrintreated nets in Dar es Salaam and Mtwara are compared in Figure 5. A slower knockdown rate was observed in Mtwara in 2004 relative to 1999 (P < 0.05) although 24-h mortality remained high. A dramatic reduction in insecticidal power occurred in Dar es Salaam over the 5-year time span: KT 50 increased from 18.2 min (95% CI: ) in 1999 to an inestimable length of time in 2004 (P < 0.05). An increase of the WHO diagnostic concentrations of pyrethroids in 2000 confounded the trends in mosquito susceptibility between 1999 and Nonetheless, between 1999 and 2004 there was an apparent increase in susceptibility in several sites as would be expected given the higher test dosage (Table 1). However, in Dodoma, the susceptibility of Cx. quinquefasciatus to permethrin and deltamethrin declined despite exposure to higher insecticide concentrations. The same occurred for Cx. quinquefasciatus in Dar es Salaam with respect to deltamethrin. These results highlight areas where resistance may be developing. In the Kilombero district of Tanzania, where the KINET project operated from 1996 to 2000, we observed 82% (95% CI: 66 98) susceptibility to alphacypermethrintreated nets in bio-efficacy tests and 97% (95% CI: ) susceptibility of An. gambiae s.l. to permethrin in susceptibility tests in Data from bio-efficacy tests in 1066 ª 2007 Blackwell Publishing Ltd

7 Table 1 Susceptibility of Tanzanian populations of malaria vectors, Anopheles gambiae s.l. and An. funestus s.l., and Culex quinquefasciatus mosquitoes to diagnostic concentrations of pyrethroids and DDT in 1999 and Mean percentage mortality in WHO susceptibility tests and number of mosquitoes exposed (n) à Species Insecticide Site (95% CI)* n (95% CI)* n Anopheles spp. Deltamethrin Dodoma Mtwara Morogoro Ifakara Mwanza Tabora Mbeya Arumeru Moshi Permethrin Dodoma Mtwara Morogoro 87 (88 93) Ifakara (93 100) 50 Tanga Mwanza Tabora Mbeya Arumeru 91 (90 93) 150 Moshi 96 (93 99) 104 DDT Morogoro Ifakara 99 (97 100) 120 Mwanza 98 (97 100) 79 Tabora Mbeya 99 (96 100) 122 Arumeru Moshi Culex quinquefasciatus Deltamethrin Dodoma 33 (28 37) Mtwara 75 (51 100) (73 96) 40 Morogoro 38 (35 50) (54 89) 75 Dar es Salaam 75 (65 85) (26 66) 55 Tanga 49 (28 69) 101 Bagamoyo 59 (55 63) 37 Ifakara 70 (63 77) 160 Moshi 76 (64 88) 111 Permethrin Dodoma 20 (13 22) Mtwara 65 (50 85) 60 Morogoro (33 53) 144 Dar es Salaam 40 (20 60) (44 51) 40 Tanga 74 (58 90) 65 Ifakara 67 (38 96) 160 Moshi DDT Dar es Salaam 4 75 Moshi 8 (5 11) 121 *Confidence intervals not applicable where <2 replicates or no between-replicate variation WHO diagnostic concentrations: deltamethrin 0.025%, permethrin 0.25%. à2004 WHO diagnostic concentrations: deltamethrin 0.05%, permethrin 0.75%, DDT (4.0%) and susceptibility tests carried out in (while ITN coverage was increasing) were compared to the 2004 data to detect changes over time. Bio-efficacy tests from Njagi village showed 100% susceptibility of An. gambiae s.l. to deltamethrin in 1999 while susceptibility to alphacypermethrin was 82% in 2004; however, ª 2007 Blackwell Publishing Ltd 1067

8 Table 2 Bio-efficacy of nets treated with deltamethrin (KO-tab TM,25mg m 2 ) and alphacypermethrin (Fendona TM,40mg m 2 ) against Tanzanian populations of malaria vectors, Anopheles gambiae s.l. and An. funestus s.l., and Culex quinquefasciatus. Mean percentage mortality and number of mosquitoes exposed (n) are presented for sites visited during surveys commissioned by PSI in 1999 and Species Insecticide Site (95% CI)* n (95% CI)* n Anopheles spp. Deltamethrin Dodoma Mtwara Ifakara Alphacypermethrin Dodoma Mtwara 94 (85 100) 24 Morogoro Ifakara 82 (66 98) 50 Culex quinquefasciatus Deltamethrin Dodoma 7 30 Mtwara (82 100) 68 Dar es Salaam 92 (85 100) 50 8 (0 16) 30 Bagamoyo 68 (50 85) 40 Alphacypermethrin Dodoma 2 (0 6) 30 Mtwara 39(24 53) 68 Morogoro 27 (8 47) 48 Ifakara 73 (65 81) 50 Dar es Salaam 17 (8 26) 30 Bagamoyo 20 (8 32) 40 *Confidence intervals not applicable where <2 replicates or no between-replicate variation. wide between-replicate variation in mortality was observed in 2004 (range %) that may reflect uneven distribution of insecticide on the net. Susceptibility tests carried out in Njagi village showed a progressive shift in knockdown time between 1997 and 1999 for permethrin and deltamethrin (Figure 6; Table 3); however, 24-h mortality was consistently 100% for both pyrethroids. In 2004 this knockdown rate returned to that seen in 1997 (data for deltamethrin only), probably due to the increased diagnostic concentrations. In permethrin tests, a slight decrease in 24-h mortality from 100% in 2000 to 97% in 2004 was observed despite the increased diagnostic concentration. In the Tanga region, in a village where ITN coverage was maintained at 70 90% between 1987 and 2004, % susceptibility of Anopheles spp. to permethrin was observed in 1999 and in Discussion This study demonstrates the continued efficacy of two pyrethroids against malaria vectors in areas of Tanzania where ITNs have been used for up to 17 years. At recommended treatment dosages, both deltamethrin- and alphacypermethrin-treated nets had strong knockdown and lethal effects on the major malaria vectors An. gambiae s.l. and An. funestus s.l. These results are promising given (a) (b) Mtwara 0 min 3 min 20 min 30 min 60 min 24 h Dar es Salaam 0 min 3 min 20 min 30 min 60 min 24 h Figure 5 Mortality of Culex quinquefasciatus in (a) Mtwara and (b) Dar es Salaam after 3-min exposure to deltamethrin-treated net (25 mg m 2 ) in ball tests carried out in 1999 (open columns) and 2004 (shaded columns). Error bars represent 95% confidence intervals ª 2007 Blackwell Publishing Ltd

9 (a) (b) 100 Deltamethrin Minutes Permethrin Minutes Figure 6 Mortality of Anopheles spp. during 1-h exposure to (a) Deltamethrin and (b) Permethrin in WHO susceptibility tests, and 2004, Njagi village, Kilombero district, Tanzania, showing shift in knockdown time. Diagnostic concentrations: , deltamethrin (0.025%) and permethrin (0.25%); 2004, deltamethrin (0.05%) and permethrin (0.75%). Table 3 Time for 50% knockdown (KT 50 in minutes, 95% CI) of Anopheles gambiae s.l. during 1-h exposure to WHO diagnostic concentrations of permethrin and deltamethrin in Njagi village, Kilombero district, Tanzania, in and 2004 Year Permethrin (0.25%) Deltamethrin (0.025%) 1997 n a 7.4 ( ) ( ) 13.9 ( ) ( ) 19.0 ( ) 2004* 9.7 ( ) 7.0 ( ) *Diagnostic concentrations in 2004: permethrin (0.75%) and deltamethrin (0.05%). present malaria control strategies to increase ITN use in Tanzania (Magesa et al. 2005), and suggest that, with high ITN coverage, current insecticide treatments on ITNs can provide an effective means of malaria control. While the high susceptibility of malaria vector mosquitoes to diagnostic concentrations of pyrethroids is encouraging, in several areas less than 100% mortality to permethrin was observed. Continued monitoring is therefore required, particularly in the areas where marginal (i.e %) susceptibility is seen. It should be noted that two of the three areas where marginal susceptibility was observed are situated in regions with large-scale agricultural production, e.g. floriculture in Arumeru region and irrigated rice production in Moshi district, so the potential contribution of agricultural insecticide use to resistance development must be considered despite lack of evidence of resistance selection in areas where pyrethroids were used for cotton production in the 1990s (Curtis et al. 1999). In the third area with marginal susceptibility, in southern Tanzania, historical data suggest that while full susceptibility was observed during years of ITN scaling-up, there was a concurrent shift in knockdown time. While we do not have genetic measures of resistance frequency, such a shift has been found to correlate with early stages of resistance development in vector populations (Chandre et al. 2000), and thus this area should be closely monitored. Interestingly, in the Tanga region of coastal Tanzania where ITNs have been distributed by research projects achieving high levels of coverage, similar high levels of susceptibility of Anopheles mosquitoes to pyrethroids have been observed. Mortality of malaria vectors after exposure to DDT was greater than 97% indicating high susceptibility to this organochlorine insecticide. While marginally lower (95%) anopheline susceptibility to DDT was observed in Tabora, the number of vectors tested was insufficient to draw conclusions. High levels of resistance to DDT were observed in Cx. quinquefasciatus in the two areas where this species was tested. Despite efforts to reduce variability in the results, several factors may have impacted on the precision of insecticide susceptibility and ITN bio-efficacy measurement. These factors include the variability between replicates associated with pyrethroid irritability and or variable insecticide distribution on individual nets, the small numbers of mosquitoes tested in some locations, and the use of wild-caught mosquetoes (i.e. non-agestandardized). Pyrethroid resistance declines with increasing mosquito age (Hodjati & Curtis 1999), therefore the use of wild-caught mosquitoes could result in variability in the response to insecticides. However, these assays were representative of the biting mosquito population that nets are meant to protect against. The issue of numbers of mosquitoes tested may be overcome by more intensive and regular sampling of mosquito populations to account for seasonal changes in population density. In addition, testing of F1 progeny of field collected mosquitoes would likely produce more consistent results. However, this would require a field insectary or laboratory for rearing of larvae. In locations where such facilities exist continued testing should be ª 2007 Blackwell Publishing Ltd 1069

10 encouraged and such studies could be enhanced by molecular analysis. Using standard bioassay methods we have found no evidence that high coverage with ITNs has selected for physiological resistance in malaria vector populations in Tanzania. Caution may be required, however, since the levels of coverage achieved in ITN projects may not have put sufficient pressure on the overall vector population to give evidence of resistance being selected. While the standing crop of nets in Tanzania is increasing the actual level of ITN coverage may be lower due to lack of regular insecticide re-treatment. A recent innovation in ITNs is the development of longlasting insecticide-treated nets (LLIN) (Guillet et al. 2001a). These wash-resistant nets eliminate the need for regular re-treatment, which has been a major barrier to maintaining sustained net efficacy in many areas (Schellenberg et al. 2002). Gradual replacement of untreated and conventionally-treated nets with long-lasting versions through increasing availability of long-lasting insecticide treatments and factory-treated LLIN will ensure wider coverage in the future. The low levels of susceptibility of Cx. quinquefasciatus, and the low efficacy of treated-nets against this widespread, nuisance-biting species must be taken into account. While the reduced insecticide efficacy against culicine mosquitoes does not directly impact on the level of malaria control achieved by ITNs, net users often value the reduction in biting by nuisance insects over perceived health benefits (Zimicki 1996; Chandre et al. 1998; Schellenberg et al. 1999; Myamba et al. 2002); therefore, reduced public acceptance may have an indirect impact on ITN effectiveness. Our data suggest that certain pyrethroids may be more insecticidal against Cx. quinquefasciatus than others, as was observed in Mtwara and Bagamoyo; however the decline in efficacy of certain insecticides over time indicates the potential for rapid resistance development in this species. In addition to an innate tolerance to residual insecticide deposits (Brown & Pal 1973), resistance to pyrethroids has been widely recorded in Cx. quinquefasciatus (Khayrandish & Wood 1993; Chandre et al. 1998), often resulting in low efficacy of ITNs against this species in experimental huts (Curtis et al. 1996; Guillet et al. 2001b) and community-based trials (Maxwell et al. 1999). Other classes of insecticides have been evaluated for their potential use on ITNs, and high levels of control have been achieved with certain carbamates and organophosphates (Miller et al. 1991; Kolaczinski et al. 2000). However, safety remains a concern with these insecticide classes, and formulations with low toxicity or methods of delivery that limit human contact may be potential options alone or in combination with pyrethroid-treated nets. Combining two classes of insecticides on nets may present a method for managing resistance, by exposing mosquitoes to two insecticides with different modes of action (Curtis 1985; Guillet et al. 2001a,b; Hougard et al. 2003), a phenomenon similar to combination therapy (ACT) currently being advocated for antimalarials (Ashley & White 2005). Close monitoring of insecticide resistance is essential, and while current insecticide treatments have demonstrated efficacy against malaria vectors in Tanzania there is no room for complacency: efficacy against non-vector nuisance mosquito populations may also be needed to ensure public acceptability and sustained ITN use. Acknowledgements The study was conducted within the Joint Malaria Programme, a collaboration between the National Institute for Medical Research in Tanzania, Kilimanjaro Christian Medical Centre, London School of Hygiene and Tropical Medicine, and the University of Copenhagen. Population Services International (PSI), Tanzania, provided financial and logistical support for surveys undertaken in 1999 and The National Malaria Control Programme commissioned a survey in with funding from the Tanzanian Ministry of Health. KINET was funded by the Swiss Agency for Development and Co-operation and the Government of Tanzania. Community-based ITN studies in the Tanga region were supported through grants from MRC UK and WHO. MAK and MER were supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). CD is supported by a research fellowship in tropical medicine (#063516) from the Wellcome Trust. JAS is funded in part by the Gates Malaria Partnership. References Abbott WS (1925) A method of computing the effectiveness of an insecticide. Journal of Economic Entomology 18, Akogbeto M & Yakoubou S (1999) Resistance of malaria vectors to pyrethrins used for impregnating mosquito nets in Benin, West Africa. Bulletin de la Societe de Pathologie Exotique 92, Andreasen AR (1995) Marketing social change. Jossey-Bass, San Francisco. Ashley EA & White NJ (2005) Artemisinin-based combinations. Current Opinion in Infectious Disease 18, Awolola TS, Brooke BD, Hunt RH & Coetzee M (2002) Resistance of the malaria vector Anopheles gambiae s.s. to pyrethroid insecticides, in south-western Nigeria. Annals of Tropical Medicine and Parasitology 96, ª 2007 Blackwell Publishing Ltd

11 Brown AWA & Pal R (1971) Insecticide Resistance in Arthropods, 2nd Edn. World Health Organization Monograph no. 38. World Health Organization, Geneva, pp Chandre F, Darriet F, Darder M et al. (1998) Pyrethroid resistance in Culex quinquefasciatus from West Africa. Medical and Veterinary Entomology 12, Chandre F, Darriet F, Manga L et al. (1999) Status of pyrethroid resistance in Anopheles gambiae sensu lato. Bulletin of the World Health Organization 77, Chandre F, Darriet F, Duchon S et al. (2000) Modifications of pyrethroid effects associated with kdr mutation in Anopheles gambiae. Medical and Veterinary Entomology 14, Corbel V, Chandre F, Brengues C et al. (2004) Dosage-dependent effects of permethrin-treated nets on the behaviour of Anopheles gambiae and selection of pyrethroid resistance. Malaria Journal 3, 22. Curtis CF (1985) Theoretical models of the use of insecticide mixtures for management of resistance. Bulletin of Entomological Research 75, Curtis CF, Myamba J & Wilkes TJ (1996) Comparison of different insecticides and fabrics for anti-mosquito bednets and curtains. Medical and Veterinary Entomology 10, Curtis CF, Miller JE, Hodjati MH, Kolaczinski JH & Kasumba I (1998) Can anything be done to maintain the effectiveness of pyrethroid-impregnated bednets against malaria vectors? Philosophical Transactions of the Royal Society of London B 353, Curtis CF, Muro AIS, Malenganisho WLM, Njunwa KJ & Fanello C (1999) Tests for susceptibility of malaria vectors to pyrethroids in an area of Tanzania where these insecticides are used in cotton cultivation. Abstract presented at MIM African Malaria Conference, Durban, South Africa. Diabate A, Baldet T, Chandre F et al. (2002) The role of agricultural use of insecticides in resistance to pyrethroids in Anopheles gambiae s.l. in Burkina Faso. American Journal of Tropical Medicine and Hygiene 67, Elissa N, Mouchet J, Riviere F, Meunier JY & Yao K (1993) Resistance of Anopheles gambiae s.s. to pyrethroids in Cote d Ivoire. Annales de la Societe Belge de Medicine Tropicale 73, Etang J, Manga L, Chandre F et al. (2003) Insecticide susceptibility status of Anopheles gambiae s.l. (Diptera: Culicidae) in the Republic of Cameroon. Journal of Medical Entomology 40, Finney DJ (1971) Probit analysis. Cambridge University Press, Cambridge. Guillet P, Alnwick D, Cham MK, Neira M, Zaim M & Heymann D (2001a) Long-lasting treated mosquito nets: a breakthrough in malaria prevention. Bulletin of the World Health Organization 79, 998. Guillet P, N Guessan R, Darriet F, Traore-Lamizana M, Chandre F & Carnevale P (2001b) Combined pyrethroid and carbamate two-in-one treated mosquito nets: field efficacy against pyrethroid resistant Anopheles gambiae and Culex quinquefasciatus. Medical and Veterinary Entomology 15, Hanson K & Worrall E (2002) SMITN 2, Tanzania: end-of-project household survey analysis. Malaria Consortium, London, UK. Hargreaves K, Koekemoer LL, Brooke BD, Hunt RH, Mthembu J & Coetzee M (2000) Anopheles funestus resistant to pyrethroid insecticides in South Africa. Medical and Veterinary Entomology 14, Henry MC, Assi SB, Rogier C et al. (2005) Protective efficacy of lambda-cyhalothrin treated nets in Anopheles gambiae pyrethroid resistance areas of Cote d Ivoire. American Journal of Tropical Medicine and Hygiene 73, Hodjati MH & Curtis CF (1999) Evaluation of the effect of mosquito age and prior exposure to insecticide on pyrethroid tolerance in Anopheles mosquitoes (Diptera: Culicidae). Bulletin of Entomological Research 89, Hougard J-M, Corbel V, N Guessan R et al. (2003) Efficacy of mosquito nets treated with insecticide mixtures or mosaics against insecticide resistant Anopheles gambiae and Culex quinquefasciatus (Diptera: Culicidae) in Cote d Ivoire. Bulletin of Entomological Research 93, Khayrandish A & Wood RJ (1993) A multiple basis for insecticide resistance in a strain of Culex quinquefasciatus (Diptera: Culicidae) from Muheza, Tanzania, studied as resistance declined. Bulletin of Entomological Research 83, Kolaczinski JH, Fanello C, Herve J-P, Conway DJ, Carnevale P & Curtis CF (2000) Experimental and molecular genetic analysis of the impact of pyrethroid and non-pyrethroid insecticide impregnated bednets for mosquito control in an area of pyrethroid resistance. Bulletin of Entomological Research 90, Lengeler C (2004) Insecticide-treated bed nets and curtains for preventing malaria (Cochrane Review). Cochrane Database of Systematic Reviews, 2004, Issue2. Art. No.: CD DOI: / CD pub2. Magesa S, Lengeler C, desavigny D et al. (2005) Creating an enabling environment for taking insecticide treated nets to national scale: the Tanzanian experience. Malaria Journal 4, 34. Makemba AM, Winch PJ, Kamazima SR et al. (1995) Implementation of a community-based system for the sale, distribution and insecticide impregnation of mosquito nets in Bagamoyo District, Tanzania. Health Policy and Planning 10, Maxwell CA, Myamba J, Njunwa KJ, Greenwood BM & Curtis CF (1999) Comparison of bednets impregnated with different pyrethroids for their impact on mosquitoes and on re-infection with malaria after clearance of pre-existing infections with chlorproguanil-dapsone. Transactions of the Royal Society of Tropical Medicine and Hygiene 93, Maxwell CA, Rwegoshora RT, Magesa SM & Curtis CF (2006) Comparison of coverage with insecticide-treated nets in a Tanzanian town and villages where nets and insecticide are either marketed or provided free of charge. Malaria Journal 5, 44. Miller JE, Lindsay SW & Armstrong Schellenberg JRM (1991) Experimental hut trials of bednets impregnated with synthetic pyrethroids or organophosphate insecticides for mosquito control in The Gambia. Medical and Veterinary Entomology 5, ª 2007 Blackwell Publishing Ltd 1071

12 Myamba J, Maxwell CA, Asidi AN & Curtis CF (2002) Pyrethroid resistance in tropical bedbugs, Cimex hemipterus, associated with use of treated bednets. Medical and Veterinary Entomology 16, N Guessan R, Corbel V, Akogbeto M & Rowland M (2007) Reduced efficacy of insecticide-treated nets and indoor residual spraying for malaria control in pyrethroid resistance area in Benin. Emerging Infectious Diseases 13, Premji Z, Lubega P, Hamisi Y et al. (1995) Changes in malaria associated morbidity in children using insecticide treated mosquito nets in the Bagamoyo District of Coastal Tanzania. Tropical Medicine and Parasitology 46, Ranson H, Jensen B, Vulule JM, Wang X, Hemingway J & Collins FH (2000) Identification of a point mutation in the voltagegated sodium channel gene of Kenyan Anopheles gambiae associated with resistance to DDT and pyrethroids. Insect Molecular Biology 9, Schellenberg JRMA, Abdulla S, Minja H et al. (1999) KINET: a social marketing programme of treated nets and net treatment for malaria control in Tanzania, with evaluation of child health and long-term survival. Transactions of the Royal Society of Tropical Medicine and Hygiene 93, Schellenberg JRMA, Abdulla S, Nathan R et al. (2001) Effect of large-scale social marketing of insecticide-treated nets on child survival in rural Tanzania. Lancet 357, Schellenberg JA, Minja H, Mponda H et al. (2002) Re-treatment of mosquito nets with insecticide. Transactions of the Royal Society of Tropical Medicine and Hygiene 96, Stump AD, Atieli FK, Vulule JM & Besansky NJ (2004) Dynamics of the pyrethroid knockdown resistance allele in western Kenyan populations of Anopheles gambiae in response to insecticidetreated bed net trials. American Journal of Tropical Medicine and Hygiene 70, Vulule JM, Beach RF, Atieli FK, Roberts JM, Mount DL & Mwangi RW (1994) Reduced susceptibility of Anopheles gambiae to permethrin associated with the use of permethrinimpregnated bednets and curtains in Kenya. Medical and Veterinary Entomology 8, WHO (1998) Test Procedures for Insecticide Resistance Monitoring in Malaria Vectors, Bio-Efficacy and Persistence of Insecticides on Treated Surfaces. WHO, Geneva. Zimicki S (1996) Promotion in sub-saharan Africa. In: Lengeler C, Cattani J and de Savigny D (eds.) Net Gain. World Health Organization TDR, Geneva and International Development Research Centre, Ottawa, pp Corresponding Author Manisha A. Kulkarni, HealthBridge (formerly PATH Canada) Nicholas Street, Ottawa, ON Canada KIN 7B7. Tel.: ; Fax: ; mkulkarni@healthbridge.ca Efficacité des filets traités aux pyréthroïdes sur les vecteurs de la malaria et les moustiques piqueurs nuisants en Tanzanie dans les endroits où les ITNs sont utilisés à long terme objectif Mesurer la sensibilité aux pyréthroïdes de populations de vecteurs de la malaria et de moustiques piqueurs nuisants en Tanzanie et Examiner l efficacité biologique des formulations actuelles d insecticides utilisés pour le traitement des filets. méthodes Anopheles gambiae Giles s.l., An. funestus Giles s.l. et Culex quinquefasciatus Say ont été collectés au cours de trois surveillances nationales et de deux études sur les filets traités aux insecticides (ITNs) en Tanzanie. L effet d assommage et la mortalité ont été mesurés dans les essais standards de l OMS pour la sensibilité et les essais de bio-efficacité de «ball-frame». Les résultats des essais de 1999 et de 2004 ont été comparés pour déterminer des tendances dans le développement de résistance. résultats An. gambiae s.l. et An. funestus s.l. étaient fortement sensibles au perméthrine (87 à 100%) et au deltaméthrine (tous à 100%) dans les essais de l OMS en 1999 et 2004, alors que la sensibilité de Culex quinquefasciatus à ces pyréthroïdes était beaucoup faible (de 7 à 100% et 0 à 84% respectivement). L efficacité des ITNs était pareillement forte sur An. gambiae s.l. et An. funestus s.l. (82 à 100%) tandis que l efficacité sur Cx. quinquefasciatus était considérablement plus faible (2 à 100%). Il n y avait aucune indication de développement de résistance dans les populations de An. gambiae s.l. ou An. funestus s.l. sur lesquelles les ITNs ont été intensivement utilisés. Cependant, la sensibilité des moustiques piqueurs nuisants, Cx. quinquefasciatus, déclinait dans certains endroits entre 1999 et conclusion La sensibilité constante des vecteurs de la malaria aux pyréthroïdes en Tanzanie est encourageant pour le contrôle efficace de la malaria avec l utilisation des ITNs. La surveillance continue est essentielle pour assurer la détection tôtdelarésistance, en particulier dans les endroits avec une utilisation massive des insecticides en agriculture et pour la santé publique où la résistance est susceptible de se développer. La faible sensibilité répandue des moustiques Culex piqueurs nuisants souligne le soucis pour l acceptation des ITNs par les utilisateurs. mots clés Anopheles gambiae, Culex quinquefasciatus, filets traités aux insecticides, pyréthroïdes, résistance, Tanzanie 1072 ª 2007 Blackwell Publishing Ltd

13 Eficacia de las redes mosquiteras impregnadas con piretroides frente a los vectores de malaria y otros mosquitos en Tanzania, en áreas con largo uso de redes mosquiteras objetivo Medir la susceptibilidad a piretroides en poblaciones de vectores de malaria y otros mosquitos responsables de picaduras molestas en Tanzania y evaluar la eficacia biológica de los métodos actuales de formulación de insecticida utilizados para el tratamiento de redes mosquiteras. métodos Se recolectaron Anopheles gambiae Giles s.l., An. funestus Giles s.l. y Culex quinquefasciatus Say durante tres estudios nacionales y dos estudios de redes mosquiteras impregnadas (RMI) en Tanzania. El efecto Knockdown y la mortalidad fueron medidas mediante pruebas de susceptibilidad estándar de la OMS y pruebas de bioeficacia tipo ball-frame. Se compararon los resultados de las pruebas de 1999 con las de 2004, para determinar las tendencias en el desarrollo de resistencias. resultados Anopheles gambiae s.l. y An. funestus s.l. eran altamente susceptibles frente a la permetrina (rango %) y la deltametrina (consistentemente 100%) en las pruebas de la OMS en 1999 y 2004, mientras que la susceptibilidad de Culex quinquefasciatus a estos piretroides era mucho menor (rango 7 100% y 0 84%, respectivamente). La eficacia de las RMI con piretroides era similar frente a An. gambiae s.l. y An. funestus s.l. (rango %) mientras que la eficacia frente a Cx. quinquefasciatus era considerablemente menor (rango 2 100%). No había indicación de desarrollo de resistencias en poblaciones de An. gambiae s.l. o An. funestus s.l. en lugares en los que las RMI habían sido usadas extensivamente; sin embargo, la susceptibilidad de mosquitos Cx. quinquefasciatus disminuyó en algunas áreas entre 1999 y conclusión La susceptibilidad mantenida de los vectores de malaria frente a los piretroides en Tanzania es esperanzadora en lo que respecta a un posible control exitoso de la malaria con RMI. La monitorización continua es esencial para asegurar una detección temprana de resistencias, particularmente en áreas con un uso intensivo de insecticidas, bien sea agrícolas o por motivos de salud pública, en los cuales la resistencia podría desarrollarse. La baja susceptibilidad generalizada a RMI de los mosquitos Culex es preocupante, pudiendo influir en la aceptación que los usuarios tengan de las redes. palabras clave Anopheles gambiae, Culex quinquefasciatus, redes mosquiteras impreganadas con insecticida, piretroides, resistencia, Tanzania ª 2007 Blackwell Publishing Ltd 1073